近日，上海科技大学白芳教授及其团队开发出使用基于深度学习的方法发现新的GluN1/GluN3A NMDA受体抑制剂。相关论文发表在2025年5月12日出版的《中国药理学报》杂志上。

在这项研究中，研究人员使用了GeminiMol，这是一个基于深度学习的分子表示框架，包含了生物活性构象空间信息。这种方法通过参考已知的活性化合物来识别具有相似结构和生物活性构象特征的潜在命中，从而实现基于配体的虚拟筛选。利用基于Geminimol的配体筛选方法，课题组研究人员从1800万个化合物文库中发现了一种有效的GluN1/GluN3A抑制剂GM-10。

值得注意的是，GM-10具有与已知抑制剂完全不同的支架。随后的全细胞膜片钳记录证实了其活性，IC₅₀为0.98±0.13μM用于GluN1/GluN3A。其IC₅₀值为3.89±0.79，需要进一步优化以提高其选择性。μM的GluN1/GluN2A和1.03±0.21μM用于GluN1/GluN3B。这项工作强调了人工智能驱动的分子表征技术在促进支架跳跃和增强基于相似性的药物发现虚拟筛选方面的潜力。

据悉，基于配体的药物发现方法通常利用分子之间的药效团相似性来筛选潜在的活性化合物。其中，支架跳跃是一种广泛关注的基于配体的先导物识别策略，通过寻找具有相似生物活性但不同支架的抑制剂，促进临床候选药物的发现。

附：英文原文

Title: Discovery of novel GluN1/GluN3A NMDA receptor inhibitors using a deep learning-based method

Author: Wang, Shi-hang, Zeng, Yue, Yang, Hao, Tian, Si-yuan, Zhou, Yong-qi, Wang, Lin, Chen, Xue-qin, Wang, Hai-ying, Gao, Zhao-bing, Bai, Fang

Issue&Volume: 2025-05-12

Abstract: Ligand-based drug discovery methods typically utilize pharmacophore similarities among molecules to screen for potential active compounds. Among these, scaffold hopping is a widely used ligand-based lead identification strategy that facilitates clinical candidate discovery by seeking inhibitors with similar biological activity yet distinct scaffolds. In this study, we employed GeminiMol, a deep learning-based molecular representation framework that incorporates bioactive conformational space information. This approach enables ligand-based virtual screening by referencing known active compounds to identify potential hits with similar structural and bioactive conformational features. Using GeminiMol-based ligand screening method, we discovered a potent GluN1/GluN3A inhibitor, GM-10, from an 18-million-compound library. Notably, GM-10 features a completely different scaffold compared to known inhibitors. Subsequent validation using whole-cell patch-clamp recording confirmed its activity, with an IC50 of 0.98±0.13μM for GluN1/GluN3A. Further optimization is required to enhance its selectivity, as it exhibited IC50 values of 3.89±0.79μM for GluN1/GluN2A and 1.03±0.21μM for GluN1/GluN3B. This work highlights the potential of AI-driven molecular representation technologies to facilitate scaffold hopping and enhance similarity-based virtual screening for drug discovery.

DOI: 10.1038/s41401-025-01571-1

Source: https://www.nature.com/articles/s41401-025-01571-1